A5059107732- Advanced X-ray and CT Imaging
- Radiation Dose and Imaging
- Cardiac Imaging and Diagnostics
- Machine Learning and ELM
- Adversarial Robustness in Machine Learning
- Thermochemical Biomass Conversion Processes
- Anomaly Detection Techniques and Applications
- Neural Networks and Applications
- Medical Imaging Techniques and Applications
- Fault Detection and Control Systems
- Advanced Battery Technologies Research
- Extraction and Separation Processes
- Radiomics and Machine Learning in Medical Imaging
- Hydrocarbon exploration and reservoir analysis
- Biodiesel Production and Applications
- Control and Dynamics of Mobile Robots
- Traffic Prediction and Management Techniques
- Radioactive element chemistry and processing
- Hydraulic Fracturing and Reservoir Analysis
- Catalysis and Hydrodesulfurization Studies
- Evaluation Methods in Various Fields
- Biofuel production and bioconversion
- Software Testing and Debugging Techniques
- Radiology practices and education
- Machine Learning and Data Classification
University of Liverpool
2024-2025
Affiliated Hospital of Hebei University
2012-2025
Nankai University
2025
Northeastern University
2024
Jilin University
2024
Jilin Medical University
2024
University of Science and Technology of China
2024
South China University of Technology
2024
Guangzhou University
2018-2023
University of Exeter
2023
Accurate global solar radiation prediction is highly essential for related research on renewable energy sources. The cost implication and measurement expertise of emphasize that intelligence models need to be applied. On the basis long-term measured daily data, this study uses a novel regularized online sequential extreme learning machine, integrated with variable forgetting factor (FOS-ELM), predict at Bur Dedougou, in Burkina Faso region. Bayesian Information Criterion (BIC) applied build...
Non-intrusive load monitoring (NILM) is a method that provides appliance power consumption information, which will help enhance the smart grid applications. This paper proposes an end-to-end NILM for multi-event identification, alleviates challenges of setting hyper-parameters and detecting multiple events in traditional methods. In this paper, convolutional neural networks are used to extract local features target event from aggregated data sliding window. Then, multi-head self-attention...
Abstract The extreme learning machine (ELM) is a type of algorithm for training single hidden layer feedforward neural network. Randomly initializing the weight between input and threshold each neuron, matrix can be calculated by least squares method. efficient ability in ELM makes it widely applicable classification, regression, more. However, owing to some unutilized information residual, there are relatively huge prediction errors involving ELM. In this paper, deep residual compensation...
Exploring point defects in FAPbI3 and their passivation strategies is essential for optimizing the stability optoelectronic properties of perovskite solar cells. In this work, possible defect complexes with PbI2-terminated FAI-terminated structures have been explored using first-principles calculations. The results indicated that some were more stable than corresponding individual defects, including VFA + VPb VI, PbFA VPb, etc., structure, which induced obvious states within band gap. Then,...
Objective: To evaluate the effectiveness of deep learning technology based on generative adversarial networks (GANs) in reducing motion artifacts cardiac magnetic resonance (CMR) cine sequences. Methods: The training and testing datasets consisted 2000 200 pairs clear blurry images, respectively, acquired through simulated CMR These were used to establish train a GAN model. assess efficacy network mitigating artifacts, 100 images with 37 real-world encountered clinical practice selected....
Abstract Fallen leaves of landscape trees, as an emerging biomass waste, were valorized using conventional hydrothermal carbonization (HC) and microwave‐assisted (MHC) pretreatments, comparatively characterized for physicochemical properties thermal degradation kinetics. The results show that MHC is superior to HC operation, because at 200℃, the process not only gives a higher hydrochar yield (45.09 vs. 39.47 wt%) with significantly reduced energy consumption (0.63 2.74 MJ g −1 ), but also...
Abstract Amyloid fibrillation kinetics of proteins associated with neurodegenerative diseases has been extensively studied using Raman spectroscopy. The normalization factor for the spectra is crucial obtaining correct indicators, especially vibrational band intensities. Here, we compared concentration dependences between absorption at 280 nm in UV–vis spectroscopy and phenylalanine (Phe) intensity 1003 cm −1 amyloid lysozyme. former exhibits better performance as factor. Using this new...
Forecasting international iron ore is a well-known issue, BIC criterion used to select the relevant variables of price.On basis traditional extreme learning machine (ELM), regular term introduced control complexity model, and genetic algorithm (GA) regularize machine.The input-layer weight matrix hidden-layer threshold (RE-ELM) model are optimized establish BIC-based regularization (BIC-GA-RELM) price prediction increase performance RE-ELM model.The results show that BIC-GA-RELM has achieved...